The victims thrashed wildly until at last they expired, their silvery remains, pocked with open sores, littering the lower Pocomoke River in August 1997.
Investigators quickly pointed to Pfiesteria piscicida, a single-celled algae blamed for fish kills in estuaries behind North Carolina's Outer Banks in the 1980s.
Almost as quickly, state and federal agencies awarded millions in research grants in an effort to unlock the mysteries of the microbe.
Now some researchers are saying that they misidentified the culprit. The allegation has sparked a debate among high-profile scientists and heightened competition for research grants.
Maryland Department of Natural Resources scientists contend that Pfiesteria killed the fish in the Pocomoke, but two sets of scientists have advanced competing theories.
Still others have seen fish die in the presence of Pfiesteria but without the telltale lesions, and none of the researchers has been able to figure out the chemical composition of Pfiesteria's toxin.
"Pfiesteria has been blown out of proportion, at least in the Chesapeake," says Vicky S. Blazer, a fish pathologist with the U.S. Geological Survey.
Wolfgang Vogelbein of the Virginia Institute of Marine Science, in a recent speech to the Chesapeake Bay Commission, stopped just short of asserting that Pfiesteria had nothing to do with the Pocomoke fish kill.
In the December issue of the Journal of Aquatic Animal Health, Vogelbein and Blazer say that Aphanomyces invadans, a fungus responsible for fish kills in Asia, was the cause of the lesions. If anything, Pfiesteria might have provided the fungus with a way to get into the fish, the scientists say. "If Pfiesteria played a role, it was an early role," says Vogelbein, "and it may not have played a role at all."
The dispute, in which it seems that few scientists agree on anything, demonstrates the depth of the mystery surrounding the microbe and the volatility of high-profile science, with its fierce competition for grants and recognition.
Pfiesteria is a one-celled organism that researchers believe has 24 life forms, one or two of which are toxic. It lies dormant in bays and rivers until conditions are ripe -- warm, slow-moving water is overloaded with nutrients, causing algal blooms -- then turns toxic and attacks fish.
In search of clues
Researchers at the University of Maryland Biotechnology Institute say they have conducted experiments in which fish died in the presence of Pfiesteria but without telltale lesions.
"The fish kills have been associated with Pfiesteria, but there has been no direct connection," says Gerardo Vasta, a professor at UM's Center of Marine Biotechnology at the Columbus Center in Baltimore. The main problem, he says, is that scientists have not been able to figure out the chemical composition of the toxin.
John S. Ramsdell, a toxicologist at the National Oceanic and Atmospheric Administration's Marine Biotoxins Laboratory in Charleston, S.C., says it is likely that Pfiesteria produces a toxin, judging from the way fish suffer in the laboratory.
But because no one has figured out its chemical structure, "the final nail in the coffin is not there," he says.
There is an apparent link between Pfiesteria and people sickened during the outbreak on the Pocomoke River in 1997, but no proof, says Glenn Morris, lead scientist in a National Institutes of Health study of the cause of diseases associated with Pfiesteria exposure.
About two dozen people complained of memory loss and confusion, symptoms that Morris says were "linked to their exposure to the Pocomoke" when Pfiesteria was present.
In October, scientists from Virginia Commonwealth University found that Kudoa, a parasite more common to fish ponds than to the Chesapeake Bay and its tributaries, was responsible for a fish kill in the James River, despite signs that pointed to Pfiesteria. They suggested that Kudoa could have been responsible for the Pocomoke fish kills.
Joanne Burkholder, a North Carolina State University researcher who helped discover Pfiesteria, scoffs at the suggestion.
"The fish [in the James River] showed none of the field symptoms we look for," she says. "The people who reported it had never seen toxic Pfiesteria. They were trying to debunk an issue they know nothing about."
Burkholder says the Chesapeake Bay is a "fringe area" for toxic Pfiesteria outbreaks because its geographic characteristics differ from those in North Carolina.
The waters behind the Outer Banks, where the deaths of a billion fish have been attributed to Pfiesteria, don't flush as well as the Chesapeake, enabling nutrients and other pollutants to build up and grow more of the algae that Pfiesteria feeds on.
What Aphanomyces, Pfiesteria and Kudoa have in common are the conditions under which they thrive: warm, brackish water loaded with nutrients.
Much of the controversy stems from scientists in various disciplines looking at the problem from different points of view, says Don Boesch, head of the University of Maryland's Center for Environmental Science.
Burkholder, a botanist, has received a great deal of attention for her Pfiesteria research. Blazer and Vogelbein are fish pathologists whose work is not as well known.
Greg C. Garman, the Virginia Commonwealth University scientist who blamed the James River fish kill on Kudoa, is a fish biologist.
Competition for research money also comes into play. Morris' study is financed by a $6 million federal grant. Vogelbein and his colleagues have received $500,000 to study the relationship between Pfiesteria and Aphanomyces.
Which came first?
Harley Spier, a fisheries biologist for Maryland's Department of Natural Resources, calls the debate "almost a chicken and egg situation."
"Did Pfiesteria provide the opening for Aphanomyces, or Aphanomyces for Pfiesteria?" she asks. "Or did the organisms work on their own?"
Ramsdell says the only way to answer those questions is to determine the chemical makeup of the toxin, administer it to the fish and see what happens.
Publicity from the Pocomoke fish kill caused about $43 million damage to the state's seafood industry, even though a University of Maryland report showed no cases of seafood poisoning or evidence of tainted shellfish, oysters or crabs traceable to Pfiesteria.
Much has been made of the speed with which Pfiesteria attacks -- death can occur within minutes, Burkholder says -- but Vogelbein and Blazer say the sores on the fish from the Pocomoke were up to 10 days old.
"At the beginning, I assumed, like everybody else, that what killed these fish was Pfiesteria," says Blazer. "But what we consistently found was a chronic inflammatory response. Pfiesteria is supposed to be very acute, but chronic inflammatory response just doesn't work that way."
Tissue studies indicate that the fish were "trying to fight off a fungal infection," Vogelbein says. Pfiesteria's role "cannot be ruled out," but it didn't cause the lesions on the fish in the Pocomoke, he says.
Pfiesteria doesn't always kill instantly, Burkholder says. "It could be that the fish develop lesions, they get distressed and secrete something, and Pfiesteria senses it and comes up. Or it could be [that] fish have been exposed to low-level toxins and move on. But because they're not doing so well, bacteria and fungi gang up on them."
Because laboratory controls cannot be applied in the wild, "we operate on the best available circumstantial evidence," she says.
Rob Magnien, head of Maryland's Pfiesteria study team, says the differences among the scientists aren't that great, despite their competing claims.
The state has "not concluded that Pfiesteria is the cause of the lesions, but there is no strong evidence for any of the other theories," he says. "We've got to look at the full picture."
Vogelbein questions the evidence that suggests that Pfiesteria killed the fish.
"Until we see the convincing evidence, we don't buy it," he says.
What that means, says Frank Daniel, a member of the Virginia Pfiesteria Task Force, is that "the jury is definitely out on the Pfiesteria episodes."
What's to blame
The suspects in recent Chesapeake Bay fish kills.
Pfiesteria piscicida: A single-celled organism with a complex life cycle that scientists believe turns toxic in warm, slow-moving brackish water with large algal blooms and large schools of fish, particularly Atlantic menhaden. It is not a disease, but a poison that kills fish, often leaving open sores.
Aphanomyces invadans: A slow-growing fungus that thrives in warm, nutrient-enriched brackish waters and causes deep, penetrating ulcers in fish. It has caused fish kills in Thailand, Indonesia, the Philippines, Australia and Japan. Some scientists have said it was responsible for the 1997 fish kills in the Pocomoke River.
Kudoa clupeidae: A microscopic parasite that liquefies fish muscle tissue. It is usually found in fish farms where the water is enriched by high concentrations of fish waste. Scientists at Virginia Commonwealth University say it was responsible for a fish kill in the James River in October.